A seminar titled “AI-Powered Semantic Analysis of Mathematical Formulae in Documents” was hosted at the AI Research Center on November 27th, 2023 as part of the AI Seminar Series. The seminar explored the topic of machine learning for mathematical formulae with Dr. Randa Almomen, Lecturer of Software Engineering in the College of Engineering.
The seminar started by highlighting the importance of correct semantic interpretation of mathematical formulae in electronic mathematical documents as a prerequisite for advanced tasks such as search, accessibility or computational processing. Especially in advanced mathematics, the meaning of characters and symbols is highly domain dependent, and only limited information can be gained from considering individual formulae and their structures. Although many approaches have been proposed for semantic interpretation of mathematical formulae, most of them rely on the limited semantics from math representation languages whereas very few use math context as a source of information.
Dr. Randa described a novel AI-enabled approach for principal extraction of semantic information of mathematical formulae from their context in documents using machine learning. Her team utilized different supervised machine learning (SML) techniques (i.e. Linear-Chain Conditional Random Fields (CRF), Maximum Entropy (MaxEnt) and Maximum Entropy Markov Models (MEMM) combined with Rprop- and Rprop+ optimisation algorithms) to detect definitions of simple and compound mathematical expressions, thereby deriving their meaning.
The seminar also showcased the corpus that has been developed utilizing Dr. Randa’s novel approach to extract desired mathematical expressions and sub-formulae and manually annotated by two independent annotators employing a standard measure for inter-annotation agreement. The speaker also highlighted the further development of a new approach to feature representation depending on the definitions’ templates that were extracted from math documents to defeat the restraint of conventional window-based features. The seminar was attended by students enrolled in the AI and machine learning courses in the Colleges of Engineering and Science, as well as faculty in the College of Engineering.
Dr. Randa Almomen is a faculty member in the Software engineering Department of the College of Engineering at Alfaisal University and a principal investigator (PI) in Alfaisal’s AI Center. She holds a PhD in Computer Science from the University of Birmingham in the UK, a thesis titled: “Context Classification for Improved Semantic Understanding of Mathematical Formulae”, and an MSc in Computer Science with Merit from the University of Birmingham (her thesis work titled, “Abstracting Symbolic Matrices”). Prior to joining Alfaisal University, Dr. Randa held an assistant professor position at Princess Nourah bint Abdulrahman University.
Alfaisal University’s AI Seminar Series explores ways artificial intelligence (AI) can benefit society, our environment, and the economy. In weekly talks, leaders from academia, industry, and NGOs, who are at the forefront of using AI for social good, showcase AI applications that are forging positive changes in healthcare, the environment, education, technology, government and more. Like the AI Center itself, the seminar series has a broad scope, covering both new methodological contributions, ground-breaking applications, and impact on society. More information about the AI Center’s activities and scientific programs can be viewed via ai.alfaisal.edu.
